64-channel label-free fluorescence detection and single-particle counting device.

A 64-channel detection system for laser-induced fluorescence (LIF) detection at the cell level is established and applied to single event counting. Generally, fluorescence detection at the cellular level requires a dyeing label to enhance the scattered light intensity for the photodetector. However, the dyeing labels, such as fluorophores, probes, and dyes, complicate sample preparation and increase cytotoxicity in the process. Therefore, label-free detection becomes essential for in vivo research. The presented 64-channel detection system is designed for label-free detection with the ability to record feeble emissions. Two linear photodetector devices are included in the system, extending the wavelength detection range to 366-680 nm with an improved spectral resolution at an average of 4.9 nm. The performance of the system was validated by detecting unlabeled human hepatocytes (L-02) and other cell-level biologic samples. In addition, the 64-channel detection system was also used for particle counting with a quartz microfluidic chip. The counting method is based on fluorescence spectra differs from those of other devices (i.e., flow cytometry and cell-sorting equipment), which use isolated irradiance intensities.

Olive oil classification with Laser-induced fluorescence (LIF) spectra using 1-dimensional convolutional neural network and dual convolution structure model.

Laser-induced fluorescence (LIF) spectroscopy is widely used for the analysis and classification of olive oil. This paper proposes the classification of LIF data using a specific 1-dimensional convolutional neural network (1D-CNN) model, which does not require pre-processing steps such as normalisation or denoising and can be flexibly applied to massive data. However, by adding a dual convolution structure (Dual-conv) to the model, the features of the 1-dimensional spectra are more scattered within one convolution-pooling process; thus, the classification effects are improved. The models were validated through an olive oil classification experiment which contained a total of 72,000 sets of LIF spectra data, and the classification accuracy rate reached ∼99.69%. Additionally, a common classification approach, the support vector machine (SVM), was utilised for the comparison of the results. The results show that the neural networks perform better than the SVM. The Dual-conv model structure has a faster convergence speed and higher evaluation parameters than those of the 1D-CNN in the same period of iterations, without increasing the data dimension.